Download PDFOpen PDF in browserSecurity Issues in Biometric SystemsEasyChair Preprint 915014 pages•Date: October 26, 2022AbstractBiometrics is a recent discussion in computer science that is defined as the study of creating computer models. It’s significant in forensics, secure access, and government and commercial applications (such as international border crossing, making government identity proofs, etc). It is a branch of computer science that investigates a person’s correct identification. It’s a pattern recognition system that uses machine learning algorithms to process data from fingerprinting, iris scanning, retina scanning, hand geometry, face recognition, voice recognition, and odor biometrics, among other methods. Uni modal and multi-modal biometrics are the two methodologies used in biometrics. Noise, spoofing, lower accuracy, and other issues might occur when uni-modal is used. Another option is multi-modal biometrics, which solves the problem of uni-modal biometrics by providing anti-spoofing techniques that make it harder for an at tacker to infiltrate the security system. They have a look at’s predominant aim is to realize the position of deep learning in the authentication system in addition to its use within the enhancement of biometric device protection. We describe those methods and look at the constraints that hold to restrict biometrically technology’s full ability. The most critical are: developing robust authentication methods, assuring the security of enrolled templates, and protecting structures from various assaults. Keyphrases: Authentication, Biometric, identification
|